The proposed research combines biophysics, statistical signal processing, computational data analysis, and neurobiology, to address a fundamentally interdisciplinary problem: How are global brain states organized? The brain's ability to operate in distinct global states, including waking, sleep, and general anesthesia, is critical for human health and medicine. This project aims to understand how cortical networks, interacting via millisecond -scale electrical signals, self-organize over macroscopic spatial dimensions to sustain global states for minutes to hours. In humans, electro- and magnetoencephalography (EEG, MEG) and intracranial electrocorticography (ECoG) probe neural dynamics with ms-scale resolution, but interpretation is challenging due to the ambiguity of their microscopic current sources. This project will use new, computationally sophisticated analyses and realistic biophysical modeling, combined with large-scale physiological recordings, to provide insight into the nature of cortical neural activity, in particular the global organizatin of sleep and general anesthesia. The mentored postdoctoral phase will build on preliminary results from electrophysiological studies of human cortical dynamics during induction of general anesthesia. By using advanced statistical signal processing of high-density EEG recordings, this research showed that the unconscious brain during general anesthesia generates two categorically distinct types of rhythmic activity. These patterns are indistinguishable by classica power spectral methods and hence were not observed previously. These results indicate propofol general anesthesia is not a unitary state, but comprises multiple global mode. The implications of these findings will be pursued by analyzing how auditory stimulus processing is altered during each state of unconsciousness evoked by propofol general anesthesia. Through computational and statistical analysis of cortical event-related potentials this project will probe the time course of neural activity following controlled auditory events to test whether the induction and emergence from anesthesia modulate sensory processing differentially. The preliminary results obtained in these studies will lead directly to the R00 independent research. Using intracranial recordings, obtained from patients implanted with arrays of electrodes in the course of treatment for epilepsy, this study will provide the first map in humans of the fine-scale spatial organization of specific activity patterns associated with general anesthesia. The propagation of currents and magnetic flux through the multiple layers of dielectric tissue in the head of a human subject will be measured empirically. Finally, the knowledge and tools resulting from these studies of general anesthesia will be leveraged to investigate the organization of rhythmic activity in physiological sleep, specifically aiming to test a new hypothesis for the circuit under- lying sleep spindles. Together, these studies will provide an empirically validated framework for understanding the global organization of neuronal activity throughout the brain within waking and unconscious states.